| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Vascular Biology Program and Department of Surgery, and
Department of Ophthalmology, Children’s Hospital Boston and Harvard Medical School, Boston, Massachusetts, USA; and
Department of Microbiology and Immunology, Uniformed Services University of the Health Sciences, Bethesda, Maryland, USA
1Correspondence: Vascular Biology Program and Department of Surgery, Children’s Hospital Boston and Harvard Medical School, Karp Family Research Laboratories, 1 Blackfan Cr., Boston, MA 02115, USA. E-mail: taturo.udagawa{at}childrens.harvard.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: tumor • stroma • cancer • Flk-1 • VEGFR2 • angiogenesis • endothelial cells • GFP
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
or bone marrow-derived circulating endothelial precursors (2)
. Tumor vascularization is promoted by angiogenesis factors that are elaborated by the tumor cells and tumor-associated host cells such as inflammatory cells (3)
and fibroblasts (4)
. Suppression of angiogenesis using inhibitors (5)
or failure to induce angiogenesis can result in a state of tumor dormancy. Conversely, an "angiogenic switch" promoted by changes in critical host-tumor cell interactions can lead to loss of tumor dormancy and to relapse. The magnitude of neovascular responses can be modulated by genetic factors that are strain dependent (6)
, which suggests that, in turn, tumor growth is controlled in part by polymorphic genes expressed by tumor-associated host cells.
To facilitate the isolation and characterization of tumor-associated stromal cells, we bred the GFP transgene from the UBI-GFP/BL6 line (7)
into SCID mice. A ß-actin promoter-driven GFP transgenic nude mouse has been described (8)
. Mice homozygous for the ß-actin-driven transgene (9)
were abnormal, however, and died shortly after birth. In contrast, UBI-GFP/BL6 mice, which express high levels of GFP under the control of the ubiquitin promoter in all cells examined, are healthy and viable when maintained as a homozygous line. Mice derived from UBI-GFP/BL6 are ideal for cell transplantation studies due to high, uniform expression levels that are distinct for different cell types, including red blood cells (RBCs) (7)
. Since considerable strain differences have been reported with respect to angiogenesis and tumor growth (6
, 10
11
12)
, we bred the GFP transgene from UBI-GFP/BL6 into C57BL/6J-scid and C.B-17-scid mice.
These transgenic mice were used to characterize tumor-associated stromal cell components. Although Flk-1 receptor expression in the adult periphery is usually restricted to the endothelium (13)
, relatively high levels of Flk-1+ cells that coexpressed the hematopoietic marker CD45, Sca-1, and c-Kit were incorporated into tumor tissue. Recent studies have indicated that circulating endothelial progenitor cells from the bone marrow can incorporate into microvessels (2)
. We therefore examined the origin of the tumor vasculature by transplanting tumors into nontransgenic SCID mice engrafted with either bone marrow or skin from GFP transgenic donors. The GFP-positive cells from the bone marrow that incorporated into tumors did not form vessels. In contrast, tumors implanted adjacent to GFP donor skin in nontransgenic animals clearly contained GFP-positive vessels recruited from the local environment. The results demonstrate there was a significant incorporation of bone marrow-derived Flk-1+/CD45+ cells in the tumor stroma, but the majority of the vasculature was derived from the local environment.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
were generated by crossing the UBI-GFP/BL6 mouse (7)
into the C57BL/6J-scid strain obtained from Jackson Laboratories (Bar Harbor, ME, USA). Genomic DNA was isolated from tails using the Qiagen DNAeasy extraction kit. Genomic DNA containing the SCID locus was amplified by PCR using the 5' primer CGGAAAAGAATTGGTATCCAC and 3' primer GGCCCCTGCTAACTTTCTCT. The SCID mutation generates an AluI site that is not present in the wild-type genomic sequence (14)
. The PCR products were digested with AluI, separated on a 6.25% polyacrylamide-TAE gel, and visualized using Cyber green (Molecular Probes, Eugene, OR, USA). Homozygous and heterozygous transgenic animals were distinguished by measuring fluorescence intensity of RBCs on a flow cytometer. RBCs from homozygous animals were twice as bright as those from heterozygous animals. Blood was collected by retro-orbital bleeding using heparinized capillary tubes, and 20 µL of blood diluted in 0.5 mL of saline was analyzed by flow cytometry. C.B-17-GFP-scid mice were generated by backcrossing the GFP transgene from UBI-GFP/BL6 mice into C.B-17-scid mice (C.B-17/Icr-SCID/Sed-Prkdcscid, Massachusetts General Hospital, Boston, MA, USA) for 10 generations. The C.B-17-GFP-scid mice were maintained and used as GFP heterozygotes.
Bone marrow transplantation
C57BL/6J-GFP-scid donor mice were anesthetized with avertin, then killed by cervical dislocation. The GFP-positive bone marrow cells were collected by irrigating the femur and tibia with sterile saline (0.9% NaCl Injection, preservative free, Abbott Laboratories, Abbott Park, IL, USA) using a syringe and a 26 g needle. The harvested cells were passaged 10 times through a 26 g needle and larger clumps were removed by filtering cells through a 40 µm nylon mesh. The cells were washed once and resuspended in the saline at a density of 5 x 107 cells/mL. Nontransgenic C57BL/6J-scid recipient mice were irradiated with 2.5 Gy (15)
from a Cesium source, then injected intravenously with 
5 x 106 donor bone marrow-derived cells in a 100 µL volume. The bone marrow-transplanted mice were housed in sterilized micro isolator cages, and 30 mg/kg of Sulfatrim® Pediatric (Alphapharma, Baltimore, MD, USA) was added to the autoclaved drinking water to prevent infections. Blood was collected by retro-orbital bleeding of lightly isofluorane anesthetized mice using heparinized micro-hematocrit capillary tubes into collector tubes containing EDTA (MICROTAINER, Fisher Scientific, Pittsburgh, PA, USA). Engraftment was monitored by flow cytometric analysis of GFP expressing cells in the peripheral blood. Bone marrow reconstitution was complete 8 wk after transplantation (7)
.
Skin transplantation
Nontransgenic C57BL/6J-scid recipient mice and C57BL/6J-GFP-scid donor mice were anesthetized using avertin as described above. The donor skin and recipient sites were shaved and disinfected with 10% Providone-iodine plus alcohol. Donor skin and the skin at the site of engraftment were removed using a 1 cm biopsy punch (AcuPunch, Acuderm Inc.). Donor skin was placed on the prepared recipient and sutured to adjacent skin using 4–5 simple interrupted stitches (5-0 Chromic gut). Buprenorphine was administered at the end of the procedure. Bacitracin ointment was applied daily to the engrafted skin for 7 days to prevent infection and drying. Gauze was placed over the grafted skin for 4 days to prevent the mice from scratching or chewing the grafts. Successful engraftment was assessed by visualizing the GFP-positive engrafted skin by epifluorescence illumination. The donor skin was fully engrafted after 4–6 wk.
Tumor experiments
The Lewis lung carcinoma (LLC) line, which originated in the C57 strain, and the human MG63Ras tumor line (16)
were cultured in DMEM (Invitrogen Life Technologies, San Diego, CA, USA) containing 7% FBS (fetal bovine serum, Invitrogen Life Technologies), 2 mM L-glutamine, 1 mM sodium pyruvate, and 100 U/mL of penicillin-streptomycin. The cells were maintained in a 37°C, 10% CO2, humidified incubator. For in vivo tumor studies, the cells were trypsinized, washed twice in sterile saline, and resuspended in saline at a density of 2 x 107 cells/mL. One million tumor cells in a 50 µL volume were injected in the subcutaneous space of the backs of shaved mice.
All animal studies were performed according to institutional regulations in facilities approved by the American Association for Accreditation of Laboratory Animal Care and in accordance with current regulations of the United States Department of Agriculture, Department of Health and Human Services, and the National Institute of Health.
Flow cytometric analyses and antibodies
Tumors were first minced using scalpels into 3 mm pieces. Digestion buffer containing 2.4 U/mL dispase grade II (Roche, Nutley, NJ, USA), 0.3 units/mL of liberase blendzyme 3 (Roche), 20 Kunitz units/mL of DNase (Sigma, St. Louis, MO, USA), and 20 U/mL hyaluronidase (ICN, Irvine, CA, USA) was added at a rate of 1 mL of digestion buffer to 50 mg of tissue. The minced tissue was triturated 10 times with a 5 mL pipette, then incubated with shaking for 10 min at 37°C. The digest was diluted with an equal volume of medium containing 10% serum and filtered through a 40 µm nylon mesh to remove particulates. The filtered cells were centrifuged at 4000x rpm for 3 min, then washed twice with PBS. The cells were resuspended in PBS containing 5% FBS, stained with antibodies according to manufacturers directions, and analyzed by flow cytometry. Phycoerythrin (PE) or allophycocyanin (APC) conjugated antibodies specific for Flk-1, c-Kit, Sca-1, CD45, and irrelevant IgG isotype control antibodies were obtained from PharMingen (San Diego, CA, USA). Biotinylated CD11b and Gr1, reacted with avidin-PE or -APC, were used to analyze myeloid differentiation markers by flow cytometry.
Immunohistochemistry
To stain for GFP, formalin fixed, paraffin embedded tissue was cut into 4 µm sections and baked onto slides for 20 min at 65°C. The deparaffinized sections were digested with 20 µg/mL proteinase K in PBS at room temperature for 30 min. The slides were rinsed twice in water, treated with 3% hydrogen peroxide in PBS for 5 min, and washed twice in 0.1x PBS. The slides were then incubated with primary anti-GFP rabbit antiserum (Abcam, Cambridge, MA, USA) diluted x1/5000 in PBS containing 10% FBS for 1 h at room temperature in a humidified chamber. The slides were rinsed twice in 0.1x PBS and incubated with biotinylated goat anti-rabbit (Vector, Burlingame, CA, USA) diluted x1/200 in PBS containing 10% FBS. The slides were visualized using a tyramide amplification kit (NEN, Boston, MA, USA) and DAB substrate (Vector). The sections were counterstained with Gill’s hematoxylin. Microvessels were stained with an anti-CD31 rat monoclonal antibody (PharMingen) diluted to 0.7 µg/mL for 30 min at room temperature, and x1/200 dilution of mouse adsorbed, biotinylated rabbit anti-rat (Vector) for 30 min at room temperature.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
shows a cross section of a LLC tumor 900 mm3 in volume. Under a fluorescent dissecting microscope, the fluorescent signal was uniform throughout the tumor (Fig. 1A
). The architecture of the tumor-host interface was visualized by immunohistochemistry of formalin fixed, paraffin embedded tumor sections using a GFP-specific antibody. Cells expressing GFP were dispersed homogeneously throughout the tumor tissue and were usually never more than a distance of 3 to 4 cells from a tumor cell. The GFP antibody stained all host cells, including single cells in isolation as well as tumor vessels that were organized as multicellular structures (Fig. 1B
).
|
Flow cytometric analysis of fresh tumor samples clearly distinguished between the host cell and the tumor cell compartments. The GFP-positive host cell population comprised 30–40% of LLC tumor tissue; the remainder was the GFP-negative tumor cell population (Fig. 2
A). The fraction of GFP-positive cells in tumor tissue by flow cytometry was consistent with the immunohistochemical staining pattern. As illustrated in Fig. 1
, most of the GFP-positive host cells present in the tumor tissue are in contact with a tumor cell. Thus, most of the cells analyzed by flow cytometry appear to be incorporated into tumor tissue rather than in the blood vessels. The GFP expressing stromal cells were characterized using antibodies specific to cell surface markers or irrelevant isotype control antibodies. The tumor tissue was enzymatically dispersed into single cells, then stained with an antibody to Flk-1/VEGFR2, whose expression is generally restricted to endothelial cells in the adult peripheral tissue. Approximately 4–5% of the tumor tissue was-positive for Flk-1 and GFP. The majority (90–95%) of the Flk-1, GFP-positive host cells coexpressed CD45, however, indicating they were nonendothelial, hematopoietic-derived cells. Previous reports have suggested that Flk-1-positive cells from the bone marrow may represent hematopoietic precursors (17
, 18)
, so they were also stained for stem cell markers Sca-1 and c-Kit. The Sca-1 antigen was expressed on a subpopulation of the GFP-positive cells and on the GFP-negative LLC tumor cells, while the c-Kit marker was expressed on a subpopulation of GFP-positive cells but not on the tumor cells. All of the c-Kit-positive cells coexpressed Flk-1, Sca-1, and GFP (Fig. 2A
). The phenotype of a human MG63Ras (16)
osteogenic sarcoma xenograft tissue was also investigated (Fig. 2B
). We chose to analyze tumor tissue from C.B-17-GFP-scid mice, since the MG63Ras tumors grew 4-fold faster in this strain than in the C57BL/6J-scid strain (data not shown). As shown in Fig. 2
, the percentage of GFP-positive host cells in the MG63Ras tumor tissue was lower than in the LLC tumors. The phenotypic profile of the GFP-positive host cells in the MG63Ras xenograft (Fig. 2B
) and the murine LLC (Fig. 2A
) were similar, however, and both tumor tissues contained a population of Flk-1 cells that coexpressed the CD45 hematopoietic marker.
|
Several recent reports have indicated that tumor Flk-1+ endothelial cells may be bone marrow derived (2)
. We therefore examined the incorporation of GFP-positive bone marrow-derived cells into tumor vessels. The origin of the tumor vessels was investigated by implanting tumors into C57BL/6J-scid mice that were transplanted with either bone marrow or skin from GFP transgenic SCID donor mice. Since GFP expressed in donor tissue can potentially elicit an immune response, nontransgenic SCID mice were used as recipients. Eight wk after bone marrow transplantation, >90% of the peripheral blood cells in the nontransgenic SCID recipients were replaced by donor GFP-positive cells coexpressing CD45 (Fig. 3
A). Specific antibody staining with CD3, CD11b, and Gr1 showed that the donor bone marrow cells from the C57BL/6J-GFP-scid mice successfully reconstituted the peripheral blood. The CD3-positive T lymphocyte marker was detectable, but since SCID mice were used, the lymphocyte counts were 5- to 10-fold lower than in wild-type mice (0.78 ±0.13 x106/µL for SCIDS vs. 9.20 ±1.91 x106/µL for wild-type mice; not shown). Cells expressing c-Kit and Flk-1 were detected in the peripheral blood of transplanted animals at levels comparable to that of control animals.
|
After confirming successful engraftment, LLC tumor cells were implanted in the subcutaneous space of the GFP-bone marrow-reconstituted animals. After 12 days, the animals were killed and the tumor-bearing animals were examined under a fluorescence dissecting microscope. GFP-positive bone marrow-derived cells were disseminated to normal peripheral tissue including skin, bone, and intestine. An intense, GFP fluorescent signal was also observed in tumor tissue (Fig. 3B
). Immunohistochemical staining of the tumor sections using GFP antibodies revealed numerous solitary single cells (Fig. 4
A, B), but in contrast to the tumors implanted in GFP transgenic animals (Fig. 1A
), tumor vessels were not stained with the GFP antibody. A dramatically different staining pattern was observed when adjacent tumor sections from the GFP bone marrow transplanted animals were stained with CD31 to highlight the tumor vessels (Fig. 4C, D
). The CD31 antibodies stained some solitary cells, but also revealed the organized architecture of the tumor vasculature. These results suggested that few if any cells contributing to vessels were recruited from the peripheral blood.
|
To visualize the GFP-positive cells that were recruited to the tumor locally from adjacent tissue, we implanted tumor cells in the subcutaneous space of GFP-positive skin from C57BL/6J-GFP-scid mice, which was engrafted onto nontransgenic C57BL/6J-scid recipient mice (Fig. 5
A, B). Twelve days after tumor cell implantation, the tumors measured 1 cm in diameter. By epifluorescece illumination, the transplanted GFP-positive skin could be seen overlying a portion of the tumor surface. When the interior of the tumors were examined, GFP-positive cells from the engrafted tissue were observed recruited to the interior of the tumor (Fig. 5D
). On the histological sections, the transplanted GFP expressing skin and hair follicles were clearly stained using antibodies specific to GFP (Fig. 5E
). Tumor vessels were stained with the GFP antibody in the tumor tissue adjacent to the transplanted GFP-positive skin (Fig. 5F
), which is in contrast to the lack of microvascular structures stained with GFP antibodies in tumors implanted in GFP bone marrow reconstituted animals (Fig. 4A, B
). Control animals injected with saline adjacent to engrafted GFP-positive skin showed no incorporation of GFP-positive vessels into normal adjacent tissue (Fig. 5G, H
). Together these data indicate that the majority of the vessels were recruited from tissue adjacent to the tumor rather than from the peripheral blood.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
.
A transgenic C57/B6 nude mouse expressing GFP under control of the ß-actin promoter has been described. However, mice homozygous for the ß-actin promoter-driven GFP transgene (9)
were abnormal and died shortly after birth. In contrast, homozygous UBI-GFP/BL6 transgenic mice are viable and completely normal despite high levels of GFP in all tissues, including RBCs. Strain differences have been reported with regard to angiogenesis and immune response to transplanted tumors (6
, 10
11
12)
, which may explain the slow growth rate noted for tumors implanted in the nude C57/B6 transgenic mice (8)
. We have observed that tumors implanted in the C57BL/6J strain of immunodeficient mice grew more slowly than tumors implanted in the C.B-17 and Balb/cJ strains of immunodeficient mice (data not shown). This suggests that the C.B-17 strain may be more ideally suited for tumor xenograft studies than the C56BL/6J strain.
Alternatively, the differential growth of transplanted tumors in C57BL/6J- and C.B-17-scid mice may reflect relevant physiologic host-tumor cell interactions that play an important role in strain-dependent, differential susceptibility to cancer. A recent study has shown that the levels of circulating endothelial progenitor and circulating endothelial cells in the peripheral blood, measured by flow cytometry, correlated with genetically heterogeneous bFGF- or VEGF-induced angiogenesis (19)
. These cells may therefore serve as useful surrogate markers of angiogenesis, but it remains to be seen whether these surrogate markers participate directly in vessel formation in a strain-dependent manner. The availability of GFP transgenic SCID mice in different strains may prove useful in this regard.
We have used the UBI-GFP/BL6-derived immune-deficient GFP transgenic mice for examining host cells associated with transplanted tumor cells. The LLC tumor line implanted in C57BL/6J-GFP-scid mice exhibited an intense, fluorescent signal when examined under a fluorescent dissecting microscope. By flow cytometry, GFP-positive host cells comprised 30–40% of the tumor tissue. Consistent with the flow cytometric analyses, immunohistochemical staining revealed that a GFP-positive host cell was usually never more than a distance of three cells from a tumor cell. The GFP staining pattern revealed a heterogeneous population of GFP-positive host cells that were dispersed homogeneously throughout the tumor. These observations underscore the notion that tumors are not simply composed of malignant cells, but instead are intimately associated with host cells.
Flow cytometric analysis of tumors implanted in GFP mice revealed a significant population tumor-associated host cells that expressed Flk-1, Sca-1, c-Kit, and CD45. Studies by Lyden et al. reported that bone marrow-derived cells contribute to tumor vascularization (2)
. To investigate the origin of the stromal cells, we implanted tumors in nontransgenic SCID mice transplanted with bone marrow or skin from GFP+ donor SCID mice. Immunohistochemical staining of tumor tissue from the bone marrow reconstituted mice using a GFP antibody revealed that bone marrow-derived cells incorporated in tumors but were spatially isolated and lacked organization. Similar staining patterns were obtained in tumors that were implanted in immune competent and SCID mice reconstituted with wild-type GFP transgenic mice (UBI-GFP/BL6). In addition, we have not seen a difference in the neovascular response between SCID and wild-type mice in a corneal model of neovascularization (data not shown), which demonstrates that the lack of bone marrow contribution to tumor vessels was not attributable to reduced lymphocyte counts in SCID mice. The use of immune-deficient mice is advantageous in GFP tissue engraftment studies because the potential for an immune response to GFP is minimized, and the engrafted mice could be applied for the study of human tumor xenografts.
In contrast to the bone marrow transplantation experiments, tumors implanted in nontransgenic SCID mice, adjacent to GFP-positive skin from transgenic donors, incorporated GFP+ tumor vessels. The results indicate that, in our model, most of the tumor vessels were nonhematopoietic, tissue resident cells from the local environment rather than bone marrow-derived cells from the circulation. Our findings support the recent report showing that bone marrow-derived cells comprise precursor for periendothelial vascular mural cells but do not contribute significantly to tumor and cytokine induced angiogenesis (20)
.
An important issue that remains is the mechanism by which the nonhematopoietic tissue contributes to tumor neovascularization. It is well known that VEGF and its receptor Flk-1 are critical role for vasculogenesis, angiogenesis, and tumor growth (13
, 21
, 22)
. A number of other vascular specific molecules such as the angiopoietins and their receptors (23
24
25)
, as well as nonspecific molecules such as platelet-derived growth factor (26)
, TGF-ß (27)
, insulin-like growth factor II (28)
, and retinoic acid binding protein-4 (29)
, coordinate with VEGF to promote vascular development (22
, 30
, 31)
. The GFP+ bone marrow transplant experiments indicate that the cellular target(s) that participate in neovascularization in response to the growth factors likely resides predominantly in nonhematopoietic tissue, although the nature of the tissue-resident, nonhematopoietic cellular target that is incorporated into tumor vessels in response to tumor-derived angiogenesis factors is not yet clear.
A population of bone marrow-derived Flk-1+ cells that expressed Sca-1, c-Kit, and CD45 were found in transplanted murine LLC tissue in C57BL/6J-GFP-scid mice as well as human MG63Ras osteogenic sarcoma tumor tissue in C.B-17-GFP-scid mice. The proportion of Flk1+/CD45+ cells in the Lewis lung tumor tissue in vivo was higher than in normal tissue (not shown). Flk-1 expression on the bone marrow-derived cells suggests they may also be recruited or stimulated by the angiogenic factor VEGF. These cells do not appear to incorporate into vessels, but may indirectly contribute to tumor angiogenesis by elaborating additional factors or enzymes that stimulate the recruitment of endothelial cells from neighboring tissue. Bone marrow-derived Flk1+ cell may potentially stimulate tumor proliferation and invasion.
A relevant issue that might arise from these observations with regard to therapy is whether disrupting the binding of VEGF to nonendothelial, Flk-1+/CD45+ cells in the tumor stroma, using Avastin (32)
, for example, could inhibit tumor growth by a mechanism other than antagonizing VEGF signaling of endothelial cells. In addition, the presence of both c-Kit and Sca-1 suggests that the Flk-1+ cells may be cells that are capable of reconstituting the hematopoietic system. The GFP SCID transgenic mice will facilitate the isolation and characterization of tumor host stromal elements that participate in the regulation of angiogenesis and tumor growth.
|
ACKNOWLEDGMENTS |
|---|
Received for publication January 18, 2005. Accepted for publication September 26, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  Y. Zhang, A. Daquinag, D. O. Traktuev, F. Amaya-Manzanares, P. J. Simmons, K. L. March, R. Pasqualini, W. Arap, and M. G. Kolonin White Adipose Tissue Cells Are Recruited by Experimental Tumors and Promote Cancer Progression in Mouse Models Cancer Res., June 15, 2009; 69(12): 5259 - 5266. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Becker, N. Rohwer, T. Funakoshi, T. Cramer, W. Bernhardt, A. Birsner, J. Folkman, and R. J. D'Amato 2-Methoxyestradiol Inhibits Hypoxia-Inducible Factor-1{alpha} and Suppresses Growth of Lesions in a Mouse Model of Endometriosis Am. J. Pathol., February 1, 2008; 172(2): 534 - 544. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Shaked and R. S. Kerbel Antiangiogenic Strategies on Defense: On the Possibility of Blocking Rebounds by the Tumor Vasculature after Chemotherapy Cancer Res., August 1, 2007; 67(15): 7055 - 7058. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Udagawa, A. E. Birsner, M. Wood, and R. J. D'Amato Chronic Suppression of Angiogenesis following Radiation Exposure Is Independent of Hematopoietic Reconstitution Cancer Res., March 1, 2007; 67(5): 2040 - 2045. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. S. Kerbel Antiangiogenic therapy: a universal chemosensitization strategy for cancer? Science, May 26, 2006; 312(5777): 1171 - 1175. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
